TÉCNICAS E INSTRUMENTOS DE RECOLECCIÓN DE DATOS

^H-labelled cells were included in initial experim ents as a n internal stan d ard w hich would serve to com pensate for any variations betw een the sam pling lines. According to B radley and Kohn (1979) an d E vans et a l (1987) the extent of ^^C-DNA elution did not affect the elution of the % - DNA if the cell n u m b er per filter w as kept below 1.10®. My resu lts (Fig. 3.3) however show th a t u n d e r the altered lysis and eluting conditions (described in section 2.4) a n d a t a cell n u m b er of 5.10® p er filter, the elution of the ®H-DNA w as affected by the increasing rate of elution of the i^C-DNA w ith X -ray dose. T his underm in ed th e p u rp o se of th e stan d ard and its u se w as therefore abandoned. By doing th is the question of th e m anner of d a ta representation w as simplified an d th u s all d ata are sim ply presen ted as th e fraction of D N A -associated radioactivity eluted from the filter in 16 h w ith th e background elution value subtracted.

W hen the dose-response curve show n in Fig. 3.4 is com pared to the dose-response curves obtained by other w orkers u sin g n e u tra l elution at

pH 9.6 [e.g. B radley and K ohn 1979; W oods 1981; Slgdestad et al. 1987; M urray et a l 1988b), th e m odified lysis conditions employed h ere clearly resulted in greater sensitiviiy of the assay.

Figure 3.4 show s th a t w hen th e fraction of DNA eluted after 16 h at pH 9.6 is plotted a g ain st dose, a sigm oidal dose-response curve w as obtained. A sh o u ld er in th e low dose region (0-5 Gy) is followed by a lin ear increase in the fraction of DNA eluted u p to 40 Gy after w hich the curve plateau s out. The sh ap e of th e dose-response curve obtained a t pH 7.4 w as also sigm oidal (Fig. 3.5); show ing a shoulder in the 0-5 Gy region, followed by a linear increase in the fraction eluted extending u p to 60 Gy an d for doses greater th a n 60 Gy a p lateau w as observed. Sigdestad et al. (1987) and M urray et al. (1988b) reported th a t they obtained a curvilinear dose-response relationship a t pH 9.6 b u t th a t at pH 7.4 th ey obtained lin e ar curves. C loser in sp ectio n of th eir resu lts o b tain ed a t pH 7.4 however show s a sparsity of d a ta a t low doses and relative insensitivity of th e a ssa y u n d e r th e co n d itio n s th ey em ployed. O n th e o th er h an d Radford (1988), O kayasu an d Iliakis (1989) an d Prise e t a l (1989c) found th a t the dose-response a t the lower pH also exhibited a sh oulder and th u s concluded th a t th e sh o u ld er an d hence th e shape of th e dose-response curve w as not a n artefact of th e elution pH.

By em ploying m ore rigorous lysis conditions [viz. u sin g NLS rath er th a n SDS and the elevated lysis tem perature of 60 ^C) O kayasu and Iliakis (1989) found th a t th ey could elim inate th e sh o u ld er (w hich th ey th u s a ttrib u te d to incom plete d é n a tu ra tio n of th e hig h er o rd er chrom atin stru c tu re ). The d o se-resp o n se curves p resen ted h ere w ere obtained u sin g the modified lysis conditions of O kayasu and Iliakis (1989) and yet a sm all shoulder w as nevertheless still observed.

O ther th a n th e sh o u ld er in the low dose region, a ten d en cy for the dose-response curves to p lateau out a t large doses w as also observed (see

Fig. 3.5). This occurred a t doses g reater th a n 40 Gy for the assay a t pH 9 .6 an d > 60 Gy a t pH 7.4. T h is ten d en cy w as found to be m ore pro n o u n ced w hen th e elution v alu es of the ®H-labelled reference cells w ere included in the calculation of th e relative elution values, as seen in Fig. 3.2. O ther au th o rs who h ad also reported a strong tendency for the

induction curve to flatten o u t a t hig h er doses, h ad In fact m ade u se of an in tern al reference and th is m ight explain their resu lts (Prise et cd. 1989a; K elland e t el. 1988). On th e o th er h an d , since an in tern al sta n d a rd was n o t em ployed here, the p resen ce of th e p lateau w ould ra th e r seem to su g g est th a t a certain proportion of th e DNA would n o t elute from the filter. It is interesting to note th a t th is ’unelutable’ proportion of 15-20 % of th e total am ount of DNA is in d ep en d en t of pH employed. T his p lateau could be the resu lt of a portion of DNA th a t is bound to the polycarbonate filter or a portion of DNA th a t h a s retain ed its higher order stru c tu re in a DNA-protein complex and is therefore unable to pass th rough the pores in th e filter.

The fraction of DNA eluted after 16 h a t pH 7.4 is less th a n th at m easu red a t pH 9.6 for the sam e X -ray dose (Fig. 3.6), indicating th a t the tech n iq u e is som ew hat less sensitive a t the lower pH. The pH effect observed here is however not nearly a s m arked as found by other workers [e.g. E vans et a l 1987; M urray e t a l 1988a). U sing th e m ore rigorous lysis conditions of 2 % NLS an d a tem p eratu re of 60 ^C, O kayasu and Iliakis (1989) reported th a t th ey h a d succeeded in elim inating th e pH effect. Even though the modified lysis conditions of O kayasu and Iliakis (1989) were employed here, the pH effect w as still observed. One obvious difference in the protocols u sed w as th e com position of the eluting buffer:

O kayasu and Iliakis (1989) h ad followed the protocol of B radley and Kohn (1979) an d used a tris solution to elute the DNA, w hereas a TPAH elution buffer w as u sed here. E vans e t al. (1987) and M urray et a l (1988a), who

reported finding a m arked pH effect, h ad in fact also u sed a TPAH eluting buffer for the neutral elution assay . This is however a tentative correlation a n d it is n o t clear how th e com position of the eluting so lu tio n could explain the pH effect.

Two contradicting th e o rie s have been p o stu lated to explain the difference in sensitivity of th e a ssa y a t the two pH values. Radford (1988) a ttrib u te d the lower elution a t pH 7.4 to incom plete rem oval of DNA- b o u n d proteins, w hereas E v an s e t a l (1986) m aintained th a t th e greater elution a t the elevated pH of 9 .6 w as due to additional alkali-labile lesions th a t w ould not occur a t th e n o rm al intracellular pH. T here seem s to be stro n g evidence in favour of b o th th e se theories. On the one h an d . Flick e t a l (1989) have show n th a t th e increased elution a t pH 9.6 is observed for X -ray induced dam age only an d notably ab sen t in restrictio n enzyme in d u ced dam age, w hich is stro n g evidence in favour of th e alkali-labile hypothesis of Evans et a l (1987). O n th e other hand, O kayasu an d Iliakis (1989) have show n th a t b y a lterin g th e lysis co n d itio n s th ey could elim in ate th e pH effect on th e elu tio n of X -irradiated DNA, w hich is stro n g evidence in su p p o rt of R adford’s theory.

3 .4 .2 Biphasic repair kinetics

The repair curves a t b o th pH 9.6 (Fig. 3.6) and pH 7.4 (Fig.3.7) show c h aracteristic biphasic rep air kinetics, exhibiting a fast com ponent with half-tim e of 5-7 m in and a slow er com ponent w ith t i/ 2 of 100-120 min. B iphasic rep air kinetics for th e n o n -d en atu rin g filter elution assay have b een obtained by several a u th o rs a t pH 9.6 (Woods 1981; W eibezahn and C oquerelle 1981; S igdestad e t a l 1987) and a t pH 7.4 (Schw artz et a l

1987; Koval and K azm ar 1988b; Fox an d McNally 1988). The above e stim ate s of the half-tim es of rep air are in reasonable agreem ent with th o se obtained for m am m alian cells by other au th o rs u sin g the n eu tral

elu tio n tech n iq u e a t pH 9.6, n am ely th a t of 2-10 m in for th e rapid com ponent and t i/ 2 of 30-120 m in for th e slow com ponent as m easured u p to 3 h post-exposure (W eibezahn an d Coquerelle 1981; Radford 1983). At th is point it is im p o rtan t to m ention th a t the tim e tak en to irradiate th e cells to 30 Gy (~5 m in) or 50 Gy (~9 m in), a t 37 °C, is n o t insignificant w hen the n u m b ers of d sb rem aining a t repair tim es of less th a n 60 m in post-exposure are considered. As a consequence, the initial fast repair could possibly be som ew hat slower th a n would appear to be the case from the t i/ 2 estim ates of the fast com ponent given here.

At bo th pH values approxim ately 50 % of the b reak s h ad rejoined w ithin 5 m in of irradiation, w hich w ould indicate th a t the m easurem ent of th e fast repair kinetics is in d ep en d en t of th e pH. M arginally different rate s of rejoining were how ever m easured for the slow com ponent a t the two pH values; approxim ately 90 % of rep air had tak en place after an in cu b atio n tim e of 60 m in in th e pH 7.4 assay, w hereas th is level of rejoining w as only attained after 180 m in a t pH 9.6. B all-park estim ates (more d a ta p oints are needed for m ore accu rate estim ates) of the half- tim es of repair are given in Table 3.1.

pH of filter elution t i/ 2 for fast repair t i/2 for slow repair

assay com ponent com ponent

9.6 ~ 6 m in ~ 1 2 0 m in

7.4 ~ 5 m in ~ 1 0 0 m in

Table 3.1 Ball-park estimates of the repair half-times for CHO cells after exposure to 30 Gy obtained from Figs. 3.6 and 3.7.

In co n tra st to th e d a ta presented here, Schw artz et a l (1987) and Koval an d K azm ar (1988b) reported no difference in the rates of repair as assayed a t the two pH values of 9.6 and 7.4. A lthough good reproducibility

w as o b tained in the rep air ex p erim en ts perform ed a t pH 7.4, fu rth er re p e a t experim ents are n e c e ssa ry to a scertain the significance of the difference in repair kinetics at th e two pH values.

Koval an d K azm ar (1988b) rep o rte d th a t th e rate of re p a ir m ea su re m e n t (in V79 cells) a t b o th pH 9.6 an d 7.4, depended on the choice of eluting solution. T hey detected slower repair kinetics w ith the TPAH buffer th a n w ith th e tris so lu tio n . They th u s p ro p o sed an in teractio n betw een the lesions a n d th e eluting solution, for exam ple the existence of a repair interm ediate w hich is destabilised only in th e TPAH solution. The rate of repair m easu red here using the TPAH buffer (Figs. 3.6 an d 3.7) w as faster th a n th a t m easured by Koval and Kazm ar (1988b) u sin g th e sam e eluting buffer (b u t different lysis conditions), an d show kinetics sim ilar to the tris solution resu lts of the above au th o rs (although c a u tio n sh o u ld be ad o p ted w h en com paring th e re p a ir k in etics of different cell types as th eir rep a ir ch aracteristics m ight be disparate). R epair experim ents w ould need to be perform ed u sing the tris eluting so lu tio n to definitively show no difference u n d e r th e m odified lysis conditions employed here.

The half-tim e of the slow com ponent of repair is n o t dissim ilar to th e half-tim e of dsb repair of 2-4 h m easured in E hrlich ascites tu m o u r (EAT) cells by velocity sed im en tatio n (Bryant an d B lôcher 1980), while th e t i/ 2 of th e fast com ponent is n o t unlike th a t of ssb repair in EAT or CHO cells (B ryant e t a l 1984; C o sta and B ry an t 1988). A hnstrom (Com m ent on Radford 1985) an d H utchinson (1989) have, on th e b asis of th e sim ilar repair rates, proposed th a t the fast repair com ponent w as due to ssb rejoining (assum ing th a t th e presence of ssb could affect th e elution p ro p erties of DNA u n d e r n o n -d en a tu rin g conditions). W eibezahn and C oquerelle (1981) an d W oods (1981) h ad p u t forw ard a n alternative

ex p lan atio n of the b iphasic rep air k inetics of th e n e u tra l elution assay. T hey p roposed th a t th e rap id co m p o n en t re p re se n te d a process of

ligation and suggested th a t th e slow com ponent reflected a type of repair th a t req u ired a m ore com plex p ro cess, possibly recom bination. This in te rp retatio n suggests th a t the b ip h asic kinetic reflects two different m echanism s of repair w hich could in tu rn suggest th a t th e n eu tral elution assay detects a t least two types of lesions or dsb.

W hen com paring Figs. 3.6 and 3.8 a discrepancy arises in the results of th e long (0.5-8 h) an d sh o rt (5-180 m in) term rep air experim ents. From Fig. 3.6 one w ould expect approxim ately 90 % of the dsb to be rejoined by 3 h post-irradiation, w hereas Fig. 3.8 show s only 80 % repair a t 3 h. A lthough it is difficult to ju d g e the significance of th is difference, considering the errors on th e m ean values vary betw een 1.5 and 8 %, it is possible th a t the slow er rate of rep air detected after 8 h w as due to the higher dose used (50 Gy vs 30 Gy). Possible dose dependence of the ti/% of th e slow com ponent of rep air w ould su p p o rt the satu rab le repair model proposed by G oodhead (1985) and W heeler (1987), b u t Radford (1987a) an d Sw iegert et a l (1989) have reported finding no dose effect in repair experim ents following exposure in th e 20 to 50 Gy dose range.

Investigation of th e rep air k in etics over longer in cu b atio n tim es (Fig. 3.8) in terestin g ly revealed th e possible presen ce of a very slow com ponent of repair w ith a half-tim e of betw een 4 an d 6 h o u rs. The re p a ir w ould need to be followed over even longer in cu b atio n tim es to asce rtain w hether th is rep resen ts a th ird repair com ponent or w hether it is ju s t a n extension of th e know n slow com ponent. Rowley an d Kort (1988) u sin g the n e u tra l elu tio n a ssa y found a c o n tin u al decrease in n u m b ers of dsb u p to 6 h post-irrad iatio n (after 20 Gy) w hich is further evidence th a t dsb rep air is n o t com plete w ithin 2-3 h a s w ould appear to